CN104350202A - Closed-loop alignment identification with adaptive probing signal design technique for web manufacturing processing systems - Google Patents
Closed-loop alignment identification with adaptive probing signal design technique for web manufacturing processing systems Download PDFInfo
- Publication number
- CN104350202A CN104350202A CN201380028046.8A CN201380028046A CN104350202A CN 104350202 A CN104350202 A CN 104350202A CN 201380028046 A CN201380028046 A CN 201380028046A CN 104350202 A CN104350202 A CN 104350202A
- Authority
- CN
- China
- Prior art keywords
- detectable signal
- actuator
- collision
- design
- coiled material
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65H—HANDLING THIN OR FILAMENTARY MATERIAL, e.g. SHEETS, WEBS, CABLES
- B65H23/00—Registering, tensioning, smoothing or guiding webs
- B65H23/02—Registering, tensioning, smoothing or guiding webs transversely
- B65H23/032—Controlling transverse register of web
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65H—HANDLING THIN OR FILAMENTARY MATERIAL, e.g. SHEETS, WEBS, CABLES
- B65H23/00—Registering, tensioning, smoothing or guiding webs
- B65H23/04—Registering, tensioning, smoothing or guiding webs longitudinally
- B65H23/18—Registering, tensioning, smoothing or guiding webs longitudinally by controlling or regulating the web-advancing mechanism, e.g. mechanism acting on the running web
- B65H23/188—Registering, tensioning, smoothing or guiding webs longitudinally by controlling or regulating the web-advancing mechanism, e.g. mechanism acting on the running web in connection with running-web
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B21/00—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant
- G01B21/22—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring angles or tapers; for testing the alignment of axes
- G01B21/24—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring angles or tapers; for testing the alignment of axes for testing alignment of axes
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65H—HANDLING THIN OR FILAMENTARY MATERIAL, e.g. SHEETS, WEBS, CABLES
- B65H2557/00—Means for control not provided for in groups B65H2551/00 - B65H2555/00
- B65H2557/60—Details of processes or procedures
- B65H2557/61—Details of processes or procedures for calibrating
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing Of Short-Circuits, Discontinuities, Leakage, Or Incorrect Line Connections (AREA)
- Paper (AREA)
- Tests Of Electronic Circuits (AREA)
Abstract
A method includes designing (308, 400, 500) probing signals for testing an alignment of actuators (202) in a web manufacturing or processing system (100) with measurements of a web (108) of material being manufactured or processed by the system. The method also includes providing (310) the probing signals during alignment testing to identify the alignment of the actuators with the measurements of the web. Designing the probing signals includes designing the probing signals based on both spatial and dynamic characteristics of the web manufacturing or processing system.
Description
Technical field
Disclosure relate generally to control system.More specifically, the disclosure relates to and utilizes the closed loop for the self-adaptive detection Design of Signal technology of coiled material manufacture or treatment system to aim to identify.
Background technology
Sheet material or other coiled materials of material is used in every way in various industry.These materials can comprise paper, couch board and with the manufacture of long paper material and other products be collected on spool.Coiled material manufacture or treatment system generally include multiple actuators of the coiled material location crossing over manufactured or process.The direction crossing over web width is called crisscross, and is called machine direction along the direction of web length.
Crisscross (CD) aims at the spatial relationship referred between center that the center in the region of each CD actuator and downstream CD web qualities measure.It is usually used to important parameter web manufacturing processes being carried out to modeling, and the performance of its impact feedback usually CD controller.
Conventional CD aims at recognition technology and usually depends on open loop " collision (bump) " or " stride " test.At these test periods, the operation of feedback CD controller is stopped, and multiple actuator by disturbance to new position.Actuator setpoint distribution (profile), web qualities measure distribution or other data values are sent to Software tool, and this Software tool runs open loop recognizer, and this open loop recognizer returns the linear or nonlinear function aimed at by existing CD and quantize.
Regrettably, turn on and off feedback CD controller and interrupted the normal operating of system, and usually there is not the control to web qualities during open-loop test.As a result, open-loop test is not usually preferred in practice.Further, many coiled material manufactures or treatment system operator usually depended on the manual observation to the misalignment sign in coiled material before starting to aim at identification test.
Summary of the invention
Present disclose provides and utilize the closed loop for the self-adaptive detection Design of Signal technology of coiled material manufacture or treatment system to aim to identify.
In a first embodiment, a kind of method comprise design detectable signal in case utilize to be manufactured by coiled material manufactured by treatment system or the measurement result of the coiled material of the material of process to test the aligning of the actuator in this coiled material manufacture or treatment system.The method is also included in aligning test period provides detectable signal to utilize the measurement result of coiled material to identify the aligning of actuator.Design detectable signal comprises and designs detectable signal based on the space of coiled material manufacture or treatment system and dynamic characteristic.
In a second embodiment, a kind of equipment comprises at least one processing unit, its be configured to design detectable signal in case utilize to be manufactured by coiled material manufactured by treatment system or the measurement result of the coiled material of the material of process to test the aligning of the actuator in this coiled material manufacture or treatment system.This equipment also comprises at least one interface, and it is configured to provide detectable signal to utilize the measurement result of coiled material to identify the aligning of actuator at aligning test period.At least one processing unit described is configured to design detectable signal based on the space of coiled material manufacture or treatment system and dynamic characteristic.
In the third embodiment, a kind of system comprises self adaptation aligned units, its be configured to design detectable signal in case utilize to be manufactured by coiled material manufactured by treatment system or the measurement result of the coiled material of the material of process to test the aligning of the actuator in this coiled material manufacture or treatment system.This system also comprises signal generator, and it is configured to provide detectable signal to utilize the measurement result of coiled material to identify the aligning of actuator at aligning test period.This self adaptation aligned units is configured to design detectable signal based on the space of coiled material manufacture or treatment system and dynamic characteristic.
According to the following drawings, description and claim, other technologies feature can be apparent for a person skilled in the art.
Accompanying drawing explanation
In order to the comprehend disclosure, now reference is carried out to the following description carried out by reference to the accompanying drawings, in the drawing:
Fig. 1 illustrates according to example coiled material of the present disclosure manufacture or treatment system;
Fig. 2 illustrates according to of the present disclosure for utilizing the closed loop of self-adaptive detection Design of Signal technology to aim at the example control system identified; And
Fig. 3 to 8 illustrates according to of the present disclosure for utilizing the closed loop of self-adaptive detection Design of Signal to aim at the exemplary method identified.
Detailed description of the invention
Fig. 1 to 8 discussed below and the various embodiments being used for describing principle of the present invention in patent document are only in the illustrated manner, and should not be interpreted as by any way and limit the scope of the invention.It will be understood by those of skill in the art that and can realize principle of the present invention in the equipment of the suitable layout of any type or system.
Fig. 1 illustrates according to example coiled material of the present disclosure manufacture or treatment system 100.In this example, system 100 comprises paper machine 102, control system 104 and network 106.Paper machine 102 comprises the various parts for producing paper products (paper web (paper web) 108 namely collected at spool 100 place).Control system 104 monitors and controls the operation of paper machine 102, and it can help the quality keeping or increase the paper web 108 produced by paper machine 102.
In this example, paper machine 102 comprises at least one head box (headbox) 112, and it crosses over machine to dispense paper mass suspension on the mesh screen or grid 113 of continuous moving.The pulp suspension entering head box 112 can comprise the xylon of such as 0.2-3%, filler and/or other materials, and wherein the remainder of suspension is water.Head box 112 can comprise dilution array of actuators, and it is crossed over paper web and distribute dilution water in pulp suspension.This dilution water can be used to assist in ensuring that the paper web 108 obtained has the more uniform basis weight of crossing over paper web 108.
The drainage component array 114 of such as vacuum tank and so on removes water as much as possible to initiate the formation of sheet material 108.Steam array of actuators 116 produces vapours, and it penetrates paper web 108 and by the latent heat treatment of steam in paper web 108, thus crosses over the temperature that paper web press section increase paper web 108.Increase in temperature can allow more easily to remove remaining water from paper web 108.Again wetting sprinkle array of actuators 118 adds little water droplet (its can by air atomizing) on the surface of paper web 108.Wetting sprinkle array of actuators 118 can be used again to control the moisture distribution of paper web 108, reduce or prevent the over-drying of paper web 108, or revising any dry striped in paper web 108.
Then usually make paper web 108 through the calender with several nips (nip) reversely rotating roller.The case surface of each rod in these rollers heats by eddy-current heating array of actuators 120.Along with each roller surfaces heats partly, roller diameter is expanded partly and is therefore increased nip pressure, and this transfers to compress paper web 108 partly.Therefore can use sense heating actuators array 120 be answered to distribute to the paper controlling paper web 108 thick (thickness).The nip of calender also can be equipped with other array of actuators, such as the array of air shower or steam shower, and it can be used to the glossiness distribution or the smoothness distribution that control paper web.
Figure 1 illustrates two additional actuators 122-124.Coarse particles (thick stock) volume actuators 122 controls the uniformity entering slurry received at head box 112 place.Steam flow actuator 124 controls the heat being sent to paper web 108 from drying.Actuator 122-124 such as can represent the valve of the flow controlling slurry and steam respectively.These actuators can be used to the dry weight and the moisture that control paper web 108.
Optional feature can be used to process paper web 108 further, such as superpressure ray machine (improving the thickness of paper web, smoothness and glossiness) or one or more platform for coating (surface all to paper applies one deck coating agent to improve smoothness and the impressionability of paper web).Similarly, additional flow actuator can be used to control the ratio of dissimilar paper pulp in coarse particles and filler material and to control the amount of the various additives (such as retention agent or dyestuff) be mixed in slurry.
This represents the concise and to the point description of the paper machine 102 that can be used to a type of producing paper products.Additional detail about this type paper machine 102 is well known in the art and is not for understanding required for the disclosure.Further, this expression a kind of paper machine 102 of particular type that can use within system 100.The other machines or equipment that comprise for the production of any other or optional feature of paper products can be used.In addition, the disclosure is not limited to use together with the system for the production of paper products, but can use with the system for the treatment of paper goods or together with producing or process the system of sundry item or material (such as couch board, clamp, plastics, textiles, metal forming or metal coiled material or other or the additional materials that manufacture as moving web material or process).
In order to control paper-making process, measure serially or repeatedly one or more attributes of paper web 108.Can in the fabrication process one or the various stage place measure paper web attribute.Then this information can be used to adjust paper machine 102, such as by the various actuators in adjustment paper machine 102.This can contribute to any deviation compensating paper web attribute and expectation target, and it can contribute to the quality guaranteeing paper web 108.
As shown in fig. 1, paper machine 102 comprises one or more sensor array 126-128, and wherein each can comprise one or more sensor.Each sensor array 126-128 can measure one or more characteristics of paper web 108.Such as, each sensor array 126-128 can comprise that moisture for measuring paper web 108, basis weight, paper are thick, the sensor of coating weight, anisotropy, color, glossiness, brilliance, turbidity, fiber-wall-element model, surface characteristics (roughness of such as surface characteristics, pattern or distribution of orientations) or any other or bells and whistles.
Each sensor array 126-128 comprises any one or more structures suitably of the one or more characteristics for measuring or detect paper web 108.Sensor in sensor array 126-128 can be fixing or scanning sensor.Can cross over paper web 108 at one or several position deploy fixation of sensor, or fixation of sensor is deployed in multiple position by the full duration can crossing over paper web 108, to measure whole web width substantially.Scanning sensor group can comprise the movable sensor of any number.
Control system 104 receives measurement data from sensor array 126-128 and uses these data to control paper machine 102.Such as, control system 104 can use measurement data to adjust any actuator or the miscellaneous part of paper machine 102.Control system 104 comprises any appropriate configuration of the operation of (such as one or more computing equipment) at least partially for controlling paper machine 102.In this example, control system 104 comprises at least one treatment facility 130 and at least one memory 132, this at least one memory 132 store by one or more treatment facility 130 the instruction and data that uses, generate or collect.Control system 104 also comprises at least one network interface 134 for being undertaken communicating by one or more network, one or more networks of all Ethernets in this way of this one or more network, signal of telecommunication network or any other or additional one or more types.Control system 104 can comprise the single computing equipment with these parts, or multiple distributive computing facility can comprise the Multi-instance of these parts.
Network 106 is coupled to the various parts (such as actuator and sensor array) of control system 104 and paper machine 102.Network 106 promotes the communication between the parts of system 100.Network 106 represents any suitable network or the combination of network of the communication between the parts in promotion system 100.Network 106 such as can represent wired or wireless Ethernet, signal of telecommunication network (such as HART FOUNDATION FIELDBUS network), pneumatic control signal network or any other or additional one or more networks.
In the system 100 of figure 1, various array of actuators represents CD array of actuators, such as dilutes actuator, steam actuator, again wetting sprinkle actuator and eddy-current heating actuator.The characteristic of the paper web 108 on the control of these array of actuators is crisscross.As mentioned above, the aligning of CD actuator is usually the important parameter in various function.CD in system 100 aims at and may change every now and then due to a variety of causes (such as machine speed change, New r4 load or paper web drift).But, be usually difficult to use normal alignment recognition technology during the normal operating of paper machine 102, identify that CD aims at.In addition, conventionally only after the out-of-alignment visible signs of CD occurs, perform aligning to identify.
Present disclose provides a kind of technology for designing the detectable signal for testing the identification experimental session disturbance actuator that current C D actuator is aimed at wherein.Can design this detectable signal in response to the misalignment detecting CD actuator, this misalignment may within system 100 with continuous, intermittent or other basic generations.Once recognize the current aligning of CD actuator, then can take corrective action, such as adjustment is used for the control signal of CD actuator to compensate misalignment.The design and use of detectable signal can occur by closed-loop fashion, and it means the feedback control technology not needing disablement control systems 104.
Although Fig. 1 illustrates an example of coiled material manufacture or treatment system 100, various change can be carried out to Fig. 1.Such as, other system can be used to produce paper products or other products.And, although be shown as including the single paper machine 102 with various parts and single control system 104, but production system 100 can comprise the paper machine of any number or have other production machines of any appropriate configuration, and system 100 can comprise the control system of any number.In addition, Fig. 1 illustrates an operating environment that wherein can use CD alignment function.This function can use in any other suitable system.
Fig. 2 illustrates according to of the present disclosure for utilizing the closed loop of self-adaptive detection Design of Signal technology to aim at the example control system 104 identified.As shown in Figure 2, use aligning recognition function in conjunction with at least one CD array of actuators 202, this at least one CD array of actuators 202 can represent in any CD array of actuators as discussed above arbitrarily and can be described as actuator " bundle (beam) ".In this example, the spatial relationship between the center measured of the CD center of aiming at the region referring to each CD actuator in array 202 and the downstream CD web qualities that undertaken by one or more scanner 126-128.
In this example, control system 104 comprises CD controller 204, and it adjusts the actuator in array 202 based on feedback.This feedback can represent any proper data, such as from the measurement result of one or more scanner 126-128.CD controller 204 can revise the set point of array 202 in any appropriate manner according to the actuator types in such as array 202 and the control logic realized in CD controller 204.CD controller 204 comprises any appropriate configuration for control CD actuator.
Control system 104 also comprises for performing the parts aiming at relevant function with CD actuator.Such as, measuring signal generator 206 generates the detectable signal be used for aiming at the actuator in test period excitation array 202.Detectable signal can represent pseudo-random binary sequence (PRBS) signal or other proper signals.Detectable signal and the signal exported by controller 204 are combined, and composite signal is supplied to the actuator in array 202.Measuring signal generator 206 comprises any appropriate configuration of the signal for generating the aligning for testing CD actuator.
Measured the characteristic of paper web 108 by one or more scanner 126-128, produce one or more distribution 208.Can before the aligning test of actuator in array 202, period and catch distribution 208 afterwards.Distribution 208 identifies the value of at least one characteristic of crossing over paper web 108.
Distribution 208 is supplied to self adaptation aligned units 210.Self adaptation aligned units 210 uses distribution 208 to perform the various functions relevant with the aligning of the CD actuator in array 202.Such as, self adaptation aligned units 210 can use distribution 208 to detect this problem before alignment issues causes the visible change to paper web 108.Self adaptation aligned units 210 can also use this distribution to identify the current aligning of actuator at test period.In addition, self adaptation aligned units 210 can use distribution 208 to verify and aim at correction after completing aligning test.In addition, self adaptation aligned units 210 can identify the signal will used at aligning test period, and self adaptation aligned units 210 can provide the information of those signals of definition to measuring signal generator 206.Self adaptation aligned units 210 comprises for identifying any appropriate configuration that CD actuator is aimed at.Self adaptation aligned units 210 such as can comprise at least one computing equipment with one or more processing unit, one or more memory and one or more network interface.
In certain embodiments, self adaptation aligned units 210 realizes comprising four different self adaptation technique of alignment of aiming at correlation function.This function comprises CD and aims at supervision, detectable signal design, closed loop identification and aim at deployment.CD aims at supervision and relates to surveillance 100 to identify whether and when alignment issues (such as by using distribution 208) occurs.Detectable signal design relates to design and is aiming at PRBS or other signals of the use of test period signal-under-test generator 206.Closed loop identification relates to the current aligning of the CD actuator identified in array 202.Aim to dispose to relate to and adjust one or more models of being used by controller 204 to be taken into account by the current aligning of CD actuator, thus allow controller 204 to help to compensate alignment issues.
Self adaptation is aimed to provide and is aimed to CD the perspective and intelligent solution identified.In certain embodiments, can automatically trigger closed loop based on one or more CD performance index and aim at identification, and can automatically define this index with the determination operation of performance baseline.Further, can monitor by runnability in the backstage of CD parametric controller.In addition, closed loop test can be performed whenever misalignment being detected, closed loop model recognizer can be performed, and use procedure data can carry out assessment models.Can and if only if that CD control performance will dispose the aligning identified after deployment to real process when improving.It can be full automatic when not having user intervention that whole self adaptation aims at recognizer.
Compared with aiming at recognition methods with conventional closed loop, this self adaptation technique of alignment can have various advantage according to embodiment.Such as, it can not require to identify that controller 204 is deactivated by test period at aligning.Further, it can use PRBS signal as detectable signal at identification test period, and automatically can be determined the characteristic (such as value, duration and position) of detectable signal by detectable signal generating algorithm.Standard open recognition methods usually adopts lasting stride disturbance, is wherein determined by user by the duration of disturbance actuator zone and position.Closed loop PRBS test can cause the CD change (such as when using same magnitude and the equal length of disturbance) to the much less that web qualities is measured.In addition, automatically can trigger test by performance monitoring algorithm, and the requirement for user intervention can not be there is.
Because the identification test used during self adaptation aligning is activity closed-loop system recognizer, so can the amount that use self-adaptive detection Design of Signal to optimize the energy in the system of being injected between recognition phase be being aimed at.Such as, it can reduce or minimize the impact of identifying on CD performance while allowing the enough excitations of injection to identify for successful aligning.The value of the disturbance in detectable signal and duration determine the amount of the energy that will be injected in system.More greatly and longer detectable signal helps recognizer usually, but its may measure in distribution at CD and cause more change (CD control performance is declined).
Especially, the design of detectable signal dynamically can be considered and observe the physical constraint of actuator.These constraints can comprise minimum and maximum actuator set point value, tolerable maximum actuator moves and for the beaming limit of labial lamina (slice lip) or eddy-current heating actuator.It is useful that the distribution of PRBS detectable signal leap actuator bundle can also distribute between recognition phase in nonlinear shrinkage, the PRBS signal obtained is had in the region of high non-linearity shrinkage ratio at leap paper web 108 and has enough coverages.
In certain embodiments, self-adaptive detection Design of Signal comprises CD actuator is grouped into different clusters, and automatically generates self-adaptive detection signal mode has the constraint of serious actuator or low gain process with process.Also in self-adaptive detection Design of Signal, cohesive process is dynamic.Especially, self-adaptive detection Design of Signal can by be adapted to process physics limit and input to export dynamic mode automatically generate detectable signal.In addition, self-adaptive detection Design of Signal can process various problem, such as:
If-CD actuator bundle is subject to severe constraints, how to generate detectable signal, because usually there is the very limited space for disposing external disturbance when having the CD process of actuator bundle of severe constraints;
-how to use sef-adapting filter to be designed for and there is relatively long time constant (process for slowly) but the detectable signal with the process of relatively short time delay; And
-how coordinating the detectable signal of different actuator zone, CD is identified, and experiment has seldom the mean value of CD quality measurements or not impact (restriction machine direction upsets).Additional detail about detectable signal design is provided below.
Although Fig. 2 illustrates an example for utilizing the closed loop of self-adaptive detection Design of Signal technology to aim at the control system 104 identified, various change can be carried out to Fig. 2.Such as, the function shown in Fig. 2 divides and is only used for illustrating.By the various component combination of control system 104 or can divide again further, such as by two or more in parts 204-206 are combined into individual feature unit.As particular example, can be created on by self adaptation aligned units 210 itself detectable signal identifying that experimental session uses.
Fig. 3 to 8 illustrates according to of the present disclosure for utilizing the closed loop of self-adaptive detection Design of Signal to aim at the exemplary method identified.Fig. 3 illustrates the example for performing the group method 300 that self adaptation is aimed at.As shown in Figure 3, self adaptation alignment procedures step 302 sentence performance baseline determine start.This can comprise such as self adaptation aligned units 210 and identify the one or more distributions 208 be associated with the normal behaviour of system 100.As particular example, this can comprise self adaptation aligned units 210 be captured in the actuator in array 202 with the amount of mis-alignment of safety artificially misalignment time one or more distributions 208 of obtaining of time durations.This can provide basis for out-of-alignment following mathematics measurement.
Whether performance monitoring occurs in step 304 place, and detected in step 306 place and out-of-alignmently to determine.This can comprise such as self adaptation aligned units 210 and the distribution 208 of most recent be determined whether out-of-alignment threshold quantity to be detected compared with baseline.If no, then process is back to step 304 for further supervision.In a particular embodiment, CUSUM algorithm can be used to come execution monitoring and misalignment detection step.In U.S. Patent Publication No. 2011/0290438, describe an operable exemplary CUSUM algorithm, it is by integrally incorporated herein by reference.
If misalignment detected, then detectable signal design occurs in step 308 place.This can comprise such as design PRBS detectable signal while consideration nonlinear shrinkage option, actuator physics limit and process dynamics.Additional detail about detectable signal design is provided below.Closed loop identification test is there is in step 310 place.This can comprise the detectable signal designed during such as measuring signal generator 206 is created on previous steps, the actuator in its disturbance array 202.One or more distribution 208 is caught at identification test period.
Closed loop recognizer is performed in step 312 place.This can comprise such as self adaptation aligned units 210 and be used in and identify that the test period distribution 208 of catching is to identify the current aligning of the actuator in array 202.As particular example, recognizer can extract the response of open loop CD spatial form from closed-loop experiment data.This algorithm can be expanded to identify complete CD model, comprise response gain, width, decay and divergence.Can support that three dissimilar contractions distribute (linear contraction, parametrical nonlinearity are shunk and nonparametric is shunk) by this algorithm.Nonlinear shrinkage can distribute by suitable detectable signal method for designing, actuator physics limit and process dynamics (time delay T
dwith time constant T
p) be attached in detectable signal generation.Provided below is the details describing the use of PRBS detectable signal in both spatial domain (CD direction) and dynamic domain (machine direction).In a particular embodiment, successive ignition (such as 30 to 40 times scanning) can be passed through and perform step 310 to collect data, and then can step 312 the term of execution use this data.
Occur in step 314 place to aim to confirm.This can comprise such as self adaptation aligned units 210 take identify aim at and determine identify whether the use of aiming at improves the operation of controller 204.Can carry out aiming at the determination confirming to produce good, medium or poor determiner here in step 316 place.If this confirmation provides good or medium result, then dispose in step 318 or 320 places identify aligning.This can comprise in controller 204, such as uses institute to identify that aligning carrys out the operation of control CD actuator.If after step 320 or this confirmation provide poor result, then carry out the determination of the detectable signal test of whether executed maximum number in step 322 place.If no, then process is back to step 308 to redesign detectable signal.
Diagnosis and report is there is in step 324 place.This can comprise the information that the diagnosis of such as control system 104 and report subsystem provide the use about self adaptation alignment algorithm.General information can comprise for misalignment detection, closed loop identification and the online event log disposed.Further, when algorithm failure, subsystem can provide the theory analysis for failure, and can provide the list of possible cause to user.Such as, the reason of algorithm failure can comprise can " noise level is too high ", " actuator bundle is by severe constraints ", " (that is, variance ratio is low) of poor quality of PRBS detectable signal " or " MD upset excessive ".This diagnostic message can for try out system or malfunction elimination is useful especially.
As mentioned above, can for both generation detectable signal designs in spatial domain and dynamic domain.As a result, in certain embodiments, the detectable signal used during self adaptation is aimed at can comprise one group of two-dimensional pseudo-random binary sequence (PRBS).
Fig. 4 illustrates the illustrative methods 400 for the detectable signal design in space and dynamic domain.As shown in Figure 4, the detectable signal design in step 401 in occurrence dynamics territory.Here, white noise generator is in step 402 place generted noise signal.This can use the value of detectable signal and the initial estimation of duration and occur.Low pass filter carries out filtering in step 404 place to the output of white noise generator.This wave filter is used to help to identify the process with long-time constant and short time delay in detectable signal design.The time constant and sweep speed that are associated with the system that will test can be used to design this wave filter.PRBS generator 406 generates PRBS basic sequence in step 406 place, the initial pseudorandom sequence such as by using the input from wave filter to carry out selective value.Such as the whether enough large determination of variance ratio is carried out by exceeding threshold value in step 408 place.If not, then repeat step 402-408.
In this example, the basis signal used during closed loop test is the PRBS in time domain.Variance ratio is the auto-covariance of PRBS and the ratio of the covariance between the PRBS with different delay.This variance ratio is the performance figure of the PRBS detectable signal in Fig. 4.Usually, the detectable signal with higher variance ratio and higher excitation value can cause better CD to aim at recognition result.By default, PRBS can have all band frequency component in dynamic frequency domain, and its dynamic frequency characteristic can be very similar to Gaussian noise.For having relatively long time constant and the process of short time delay, the high frequency pumping of detectable signal can not have excessive impact to the execution of underlying process.
In order to inject higher energy at closed loop PRBS test period in process, low pass filter is attached in basis signal generation.A function of wave filter generates the binary sequence with relatively long pulse in the time domain.Generally speaking, low pass filter increase is injected into the energy in system, but reduces the value of variance ratio, and the degree of accuracy of this and recognition result is closely related.The general effect of step 402-408 generates the binary random sequences with relatively long pulse.
This method is very useful for having the process of relatively long time constant and short time delay in practice, such as usually see in the thick CD control system of paper those.As particular example, the induction CD actuator for the thick control of paper can have greatly to time constant and the low small time delay by ten seconds of 800 seconds.For this type process, low pass filter helps to perform to aim at and identifies, because utilize wave filter and the search of iteration PRBS basis signal, recognition result can improve significantly.
There is the detectable signal design in time domain in step 409, wherein, the PRBS basis signal generated during being used in step 401.In step 410 place, search is performed to candidate's actuator zone that will use at test period.This can comprise the subset such as selecting the actuator in array 202.In step 412 place, whether take into account the judgement of nonlinear shrinkage in systems in which.This can such as use the selection of user.If like this, then carry out selected actuator in step 414 place and whether can be provided for enough edges cover that accurate nonlinear shrinks identification.If not, then process is back to step 410 to select another subset of actuator.
If do not take into account nonlinear shrinkage or selected actuator to provide enough coverings for nonlinear shrinkage, then identify the detection direction being used for actuator in step 416 place.Further, in step 418 place, while considering actuator constraint, value and the symbol of each detectable signal is determined.Actuator bundle can be crossed in one-sided, bilateral or mixing side pattern, dispose detectable signal.Further, the position of detectable signal, value, direction and duration automatically can be determined when not having user to get involved.
In the spatial domain, in a random basis at actuator selected by disturbance just and/or in negative direction.In certain embodiments, the pattern of detectable signal can be depended on and actuator setpoint is switched to negative detecting location from positive detecting location, switch to neutral position from positive detecting location or switch to neutral position from negative detecting location.
The determination the need of changing for the group type of detectable signal is carried out in step 420 place.As mentioned above, actuator bundle can be crossed in one-sided, bilateral or mixing side pattern, dispose detectable signal.One-sided pattern instruction detectable signal can inject collision for all actuators in subset on a direction (up or down).Bilateral pattern instruction detectable signal can inject collision for the actuator in subset in both direction (up and down).Mixed pattern is one-sided and combination that is bilateral pattern.Can automatically realize one-sided PRBS for the actuating system of severe constraints to detect, and Kernel-based methods gain and forecasting process noise level can calculate the value of detectable signal backward.When required value exceedes the physics limit of actuator, complete the automatic switchover of the group type of PRBS detectable signal in step 418 place wherein.
Once determine the value of each PRBS signal and symbol and do not need cluster to change, then determine final PRBS signal in step 422 place.As here seen, actuator physical constraint and selected pucker options are attached in detectable signal search.In addition, detectable signal is adaptive to space and the dynamic characteristic of just controlled process.
Except the automation generation of PRBS signal, the detectable signal that self adaptation technique of alignment uses can also have other advantages.Such as, compared with testing with open loop persistence " collision ", detectable signal can be measured distribution to the CD of paper web 108 and have less impact.Reason for this point is dual.First, utilize the feedback modifiers that can be used for compensating the interference of any process to perform with closed loop and identify test.Secondly, detectable signal be one group just and/or negative pulse instead of persistence disturbance, it may have coiled material attribute and has a strong impact on.
Fig. 5 to 8 illustrates the how concrete exemplary method for detectable signal design.In the following discussion, " group type I " refers to the excitation of single actuator, " group type II " refers to the excitation of dual actuator side by side with the one-tenth group reducing detection value, and " group type III " refers to three the actuator excitations being formed as pyramid group having further value and reduce.
As shown in Figure 5, the sophisticated method 500 for PRBS Design of Signal is provided.In step 502 place design PRBS basic sequence v.This can comprise such as perform above-mentioned steps 401 to create basic sequence.Carry out in step 504 place expecting automatically or the judgement of manual designs pattern.This can comprise the setting such as using user's input or previously established.
If expectation automatic mode, then perform automatic PRBS in step 506 place and design.Figure 6 illustrates the example embodiment of this step, Fig. 6 shows how to create collision for PRBS signal.Carry out in step 508 place in PRBS signal, use one-sided, that bilateral still mixes collision determination.This can based on the factor of any number, and such as whether actuating system is subject to severe constraints.If use mixing collision, then determine one or more directions of the collision in each region in step 510 place.For one-side collision (from step 508 or 510), in step 512 place, amendment PRBS basic sequence v, makes to collide and is on suitable direction.For bilateral collision (from step 508 or 510) or after determining one-side collision, determine value and the symbol of each collision in step 514 place.Further, crashworthness is determined in step 516 place by the symbol of each collision and value being combined.Especially, this number that can comprise the normal impact attempted in maintenance detectable signal is as far as possible close to the number of the negative collision in detectable signal.
If expect manual mode in step 504 place, then perform manual PRBS in step 518 place and design.User such as can define the desired magnitude of the collision in detectable signal.In step 520 place, user's instruction will use one-sided, bilateral still to mix collision in PRBS signal.If use one-sided or mixing collision, then carry out the determination whether one-side collision can retain the value of required disturbance in step 522 place.This can comprise such as based on current actuator set point distribution and physical constraint determine whether that can adapt to user specifies detectable signal.If not, then step 526 place amendment PRBS signal ask value.If after step 524 or one-side collision can retain desired magnitude, then determine crashworthness in step 528 place by the symbol of each collision and value are carried out combination.If in the collision of step 520 place request bilateral, then whether the set point carrying out current actuator in step 524 place is in the determination in the intermediate range of the potential value of actuator.If not, then step 528 place amendment PRBS signal ask value, and determine crashworthness in step 528 place.
No matter be select automatically or manual designs, determine the PRBS impact matrix designed in advance in step 530 place.This can comprise the matrix such as defined for the actuator will tested between aligning recognition phase.Each entry in matrix can identify the crashworthness of the associated actuator be used in array 202.
In figure 6, the method 600 designed for automatic RPBS is shown.Can such as using method 600 during the step 506 of Fig. 5.As shown in Figure 6, the current cluster type being used for the PRBS signal be just designed is identified in step 602 place.If current cluster type is " type i ", then carry out in step 604 place the trial finding the first collision in proposed signal.Suppose that it exists in step 606 place, then carry out the trial finding last collision in proposed signal in step 608 place.Suppose that it exists in step 610 place, then the collision between first collision and center of step 612 place identification signal, and identify the collision between last collision and center in step 614 place.Carry out the trial of centre collision being carried out combining in step 616 place, and if success, then in step 618 place, this process terminates.In this, this process can export the collision set for " type i " single actuator excitation.
If any trial is in step 606, the failure of 610 and 618 places, then in step 620 place, group type is switched to " Type II " and reduce collision value.If be " Type II " at this place or in step 602 place group type, then carry out the trial finding first to collide side by side in step 622 place.Suppose that it exists in step 624 place, then carry out in step 626 place finding the trial of finally colliding side by side.Suppose that it exists in step 628 place, then identify the collision between the first collision and center in step 630 place, and identify the collision between last collision and center in step 632 place.Carry out the trial of centre collision being carried out combining in step 634 place, and if success, then in step 636 place, this process terminates.In this, this process can export for " Type II " side by side dual actuator excitation collision set.
If any trial is in step 624, the failure of 628 and 636 places, then in step 638 place, group type is switched to " type-iii " and reduce collision value further.If be " type-iii " at this place or in step 602 place group type, then carry out the trial finding the low edge for pyramid group to collide in step 640 place.Suppose that it exists in step 642 place, then carry out the trial finding the high rim for pyramid group to collide in step 644 place.Suppose that it exists in step 646 place, then identify the collision between the first collision and center in step 648 place, and identify the collision between last collision and center in step 650 place.Carry out the trial of centre collision being carried out combining in step 652 place, and if success, then in step 654 place, this process terminates.In this, this process can export the collision set for " type-iii " pyramidal three actuators excitation.Otherwise, generate in step 656 place and output error report, because can not determine the group of collision.
Fig. 7 illustrates the exemplary method 700 for finding low edge or high rim collision completed repeatedly in figure 6.As shown in Figure 7, carry out the trial on the border defining low and high rim in step 702 place, and carry out the trial of intermediate range zone location between low and high rim in step 704 place.This can comprise such as identify first and last collision should position near it identify the intermediate range region between those positions.If can be identified in those positions, step 706 place and exist, then identify first or finally collide in step 708 place, and in step 710 place, the collision course being used for this region is arranged to bilateral collision.
Otherwise, identify collision course in step 712 place.For one-sided or mixing collision, carry out the trial of colliding any unidirectional collision (those collisions being increased on value) in step 714 place.If in the success of step 716 place, then in step 718 place definition first or finally collide, and in step 720 place, region direction is arranged to plus or minus.
If the collision course identified in step 712 place is bilateral, then in step 722 place setting diagnostic markers, instruction can not position low edge or high rim collision.After step 722 or in response to the failure at step 716 place, identify current cluster type in step 724 place.If current cluster type is " type i " or " Type II ", then switch group type in step 726 place, and initiate repeat search in step 728 place.Otherwise if current cluster type is " type-iii ", then in the report of step 730 place generation error, and this process terminates.
Fig. 8 illustrates the exemplary method 800 for finding middle collision also completed repeatedly in figure 6.As shown in Figure 8, in step 802 place, area indices i is arranged to the exponential quantity of the first collision, and the determination within the left one side of something whether being in actuator bundle in the value that step 804 place carries out i.If like this, then in step 806 place by area indices i incremental spacing desired value, and value j is initialized as zero.The determination that whether there is bilateral collision in area indices i is carried out in step 808 place.If not, then in step 810 place by value j compared with interval target value.If value j is less than interval target value, then in step 812 place incremental zone index i and value j, and this process is back to step 808.
If really exist in step 808 place bilateral collision, then in step 814 place by value j compared with interval target value.If it is less, then in step 816 place, the collision area index being used for the current region defined by i is arranged to two (collisions of instruction bilateral), and is in current region place deployment bilateral PRBS collision in step 818.
If exceed interval target in step 810 or 814 place value j, then identify collision course in step 820 place.If use bilateral collision, then process is back to step 804.Otherwise, if use one-sided or mixing collision, then in step 822 place, value i is arranged to its currency and deducts interval target, and value j is arranged to zero.The determination that whether there is one-side collision at the current region defined by i is carried out in step 824 place.If not, then in step 826 place by value j compared with interval target value.If j is less, then increase progressively the value of i and j, and this process is back to step 824.If j is comparatively large, then process is back to step 804.
If there is one-side collision at the current region defined by i, then in step 830 place by value j compared with interval target value.If it is less, then carrying out one-side collision in step 832 place is just or negative determination.For positive one-side collision, the collision area exponential quantity being used for current region is arranged to one in step 834 place, or for negative one-side collision, the collision area exponential quantity being used for current region is arranged to negative one in step 836 place.In any one situation, be in current region place in step 838 and dispose one-sided PRBS and collide.
This process can be repeated for all actuator zones in left one side of something of actuator bundle.At certain some place, area indices i reaches Central exponent, and step 804 makes this process be redirected to right one side of something bundle PRBS search at step 840 place.Right one side of something search can be similar to the process performed in step 802-838.
Especially, the method shown in Fig. 5 to 8 supports various feature, comprising:
-aim between recognition phase with the grouping of the CD actuator zone of different cluster mode (single, side by side, pyramid);
If when-user cluster-specific type and current actuator set point distribute and/or the setting of actuator physical constraint is not suitable with the detectable signal of the type, the adjustment that group type is carried out and the correspondence adjustment that detectable signal value is carried out;
-detectable signal crosses over the deployment of mixing of CD actuator Shu Zuowei one-sided disturbance, bilateral disturbance or one-sided and bilateral disturbance;
-under each scanning identifying experiment, keeping the number of positive detectable signal as far as possible close to the trial of the number of negative detectable signal, it is chaotic that it can reduce or minimize the machine direction caused by identifying;
-in order to process the use of the sef-adapting filter in the detectable signal of the process with extensive long-time constant and relative short time delay; And
The design of the detectable signal in-spatial domain and dynamic domain.In the spatial domain, detectable signal can be specified by different detection modes (detection direction), group type and edges cover that is linear or nonlinear shrinkage distribution.In dynamic domain, PRBS basic function can be specified by different low pass filters and different variance ratios.Preferably balance between the quality that iterative search method can be used to encourage with PRBS and the energy being injected into process is to draw detectable signal.
Although the closed loop that Fig. 3 to 8 illustrates for utilizing self-adaptive detection Design of Signal to carry out aims at the example knowing method for distinguishing, various change can be carried out to Fig. 3 to 8.Such as, although be illustrated as series of steps, the various steps in every width figure can overlapping, occur concurrently, occur according to different order or occur repeatedly.
In certain embodiments, realized or support above-mentioned various function by computer program, this computer program is formed by computer readable program code and embodies in computer-readable medium.Term " computer readable program code " comprises the computer code of any type, comprises source code, object code and executable code.Term " computer-readable medium " comprises can by the medium of any type of computer access, the such as memory of read-only storage (ROM), random-access memory (ram), hard disk drive, CD (CD), digital video disks (DVD) or any other type.
The definition of setting forth some word and the term used throughout patent document may be favourable.Term " coupling " and derivative thereof refer to any direct or indirect communication between two or more elements, and no matter whether those elements carry out physical contact mutually.Term " application " and " program " refer to one or more computer program, software part, instruction set, process, function, object, class, example, related data or are suitable for its part of realization in suitable computer code (comprising source code, object code or executable code).Term " transmission ", " reception " and " communication " and its derivative are contained directly and indirect communication.Term " comprises " and " comprising " and its derivative mean comprise and do not limit.Comprising property of term "or", mean and/or.Term " with ... be associated " and " associated " and its derivative can mean to comprise, be included in ... interior, with ... interconnect, comprise, be comprised in ... interior, be connected to or with ... be connected, be coupled to or with ... coupling, can be with ... communication, with ... cooperation, staggered, side by side, close to, be bound to or with ... bind mutually, have, have ... attribute, with ... there is relation or have with ... relation, etc.
Although the disclosure has described some embodiment and the method generally associated, change and the displacement of these embodiments and method will be apparent for a person skilled in the art.Correspondingly, the above description of example embodiment does not limit or retrains the disclosure.When not departing from by spirit and scope of the present disclosure defined by the following claims, other change, replace and change is also possible.
Claims (15)
1. a method, comprising:
Design (308,400,500) detectable signal, described detectable signal be used for utilizing manufactured by coiled material treatment system (100) is manufactured or the measurement result of the coiled material (108) of the material of process to test the aligning of the actuator (202) in described system; And
There is provided (310) described detectable signal to utilize the measurement result of described coiled material to identify the aligning of described actuator at aligning test period;
Wherein, design described detectable signal to comprise and design described detectable signal based on the space of described coiled material manufacture or treatment system and dynamic characteristic.
2. the process of claim 1 wherein:
Design described detectable signal to be included in the spatial domain be associated with described system and the dynamic domain be associated with described system and to design (401,409) described detectable signal;
The width crossing over described coiled material guides described spatial domain; And
Length along described coiled material guides described dynamic domain.
3. the process of claim 1 wherein, design described detectable signal and comprise determining how in described detectable signal, to dispose disturbance, described disturbance be deployed as one-side collision, bilateral collision or one-sided collide with bilateral mix in any one.
4. the process of claim 1 wherein:
Design described detectable signal and comprise determining how in described detectable signal, to dispose disturbance, described disturbance comprises normal impact and negative collision; And
Described detectable signal is designed such that the number of normal impact is at least approximately equal to the number of negative collision.
5. the process of claim 1 wherein, design described detectable signal and comprise:
Be the first cluster by the sub-set selection of described actuator, described first cluster has the first group type;
At least one in low edge collision, high rim collision and middle collision is attempted to be positioned in basic detectable signal; And
If can not position at least one in described collision, be then the second cluster by another sub-set selection of described actuator, described second cluster has the second group type.
6. the process of claim 1 wherein, design described detectable signal and comprise:
Identify the user's cluster-specific type be associated with described detectable signal;
Determine that the distribution of current actuator set point or actuator physical constraint set up the detectable signal that whether can not adapt to described user's cluster-specific type; And
Determine based on described, adjust group type and the value of described detectable signal.
7. an equipment, comprising:
At least one processing unit (130), be configured to design detectable signal, described detectable signal be used for utilizing manufactured by coiled material treatment system (100) is manufactured or the measurement result of the coiled material (108) of the material of process to test the aligning of the actuator (202) in described system; And
At least one interface (134), is configured to provide described detectable signal to utilize the measurement result of described coiled material to identify the aligning of described actuator at aligning test period;
Wherein, at least one processing unit described is configured to design described detectable signal based on the space of described coiled material manufacture or treatment system and dynamic characteristic.
8. the equipment of claim 7, wherein, at least one processing unit described is configured to position by automatically identifying the disturbance in described detectable signal in the absence of user input, value, direction and duration and designs described detectable signal.
9. the equipment of claim 7, wherein, at least one processing unit described is configured to design described detectable signal by the following:
Be the first cluster by the sub-set selection of described actuator, described first cluster has the first group type;
At least one in low edge collision, high rim collision and middle collision is attempted to be positioned in basic detectable signal; And
If can not position at least one in described collision, be then the second cluster by another sub-set selection of described actuator, described second cluster has the second group type.
10. the equipment of claim 9, wherein, at least one processing unit described is configured to the subset distributing to select actuator based on the contraction be associated with described coiled material manufacture or treatment system.
The equipment of 11. claims 7, wherein, at least one processing unit described is configured to design described detectable signal by the following:
Low pass filter (404) is used to carry out design basis detectable signal; And
Described basic detectable signal is used to design the detectable signal comprising disturbance.
The equipment of 12. claims 11, wherein, at least one processing unit described is configured to formation base detectable signal iteratively and designs described basic detectable signal till the variance ratio be associated with one of described basic detectable signal meets threshold value.
13. 1 kinds of systems, comprising:
Self adaptation aligned units (210), be configured to design detectable signal, described detectable signal be used for utilizing manufactured by coiled material treatment system (100) is manufactured or the measurement result of the coiled material (108) of the material of process to test the aligning of the actuator (202) in described coiled material manufacture or treatment system; And
Signal generator (206), is configured to provide detectable signal to utilize the measurement result of coiled material to identify the aligning of described actuator at aligning test period;
Wherein, described self adaptation aligned units is configured to design described detectable signal based on the space of described coiled material manufacture or treatment system and dynamic characteristic.
The system of 14. claims 13, wherein, described self adaptation aligned units is configured to design described detectable signal by the following:
Be the first cluster by the sub-set selection of described actuator, described first cluster has the first group type;
At least one in low edge collision, high rim collision and middle collision is attempted to be positioned in basic detectable signal; And
If can not position at least one in described collision, be then the second cluster by another sub-set selection of described actuator, described second cluster has the second group type.
The system of 15. claims 13, wherein, described self adaptation aligned units is configured to design described detectable signal by the following:
Use low pass filter come iteratively formation base detectable signal till the variance ratio be associated with one of described basic detectable signal meets threshold value; And
Described basic detectable signal is used to design the detectable signal comprising disturbance.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US13/433,101 US9511969B2 (en) | 2012-03-28 | 2012-03-28 | Closed-loop alignment identification with adaptive probing signal design technique for web manufacturing or processing systems |
| US13/433101 | 2012-03-28 | ||
| PCT/CA2013/000241 WO2013142960A1 (en) | 2012-03-28 | 2013-03-13 | Closed-loop alignment identification with adaptive probing signal design technique for web manufacturing processing systems |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104350202A true CN104350202A (en) | 2015-02-11 |
Family
ID=49236158
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201380028046.8A Pending CN104350202A (en) | 2012-03-28 | 2013-03-13 | Closed-loop alignment identification with adaptive probing signal design technique for web manufacturing processing systems |
Country Status (5)
| Country | Link |
|---|---|
| US (1) | US9511969B2 (en) |
| EP (1) | EP2831341A4 (en) |
| CN (1) | CN104350202A (en) |
| CA (1) | CA2867212C (en) |
| WO (1) | WO2013142960A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109208373A (en) * | 2017-07-06 | 2019-01-15 | 霍尼韦尔国际公司 | For enhancing the defect analysis of continuous coiled material sheet material, classification and the reparation of equipment efficiency and output |
| CN110227294A (en) * | 2019-06-17 | 2019-09-13 | 中国工程物理研究院激光聚变研究中心 | Polishing fluid circulated filter system |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5880808B2 (en) * | 2011-04-21 | 2016-03-09 | 東洋製罐株式会社 | Bag making equipment |
| US20170045419A1 (en) * | 2015-08-10 | 2017-02-16 | Honeywell International Inc. | System and method for step test-free machine modeling using statistical information about multiple web manufacturing or processing systems |
| US10459428B2 (en) * | 2016-03-08 | 2019-10-29 | Honeywell Limited | Optimal closed-loop input design for identification of flat-sheet process models |
| US10534325B2 (en) | 2016-07-14 | 2020-01-14 | Honeywell International Inc. | Adaptive control techniques for pH control or control of other industrial processes |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4093988A (en) * | 1976-11-08 | 1978-06-06 | General Electric Company | High speed frequency response measurement |
| US5170357A (en) * | 1989-10-31 | 1992-12-08 | Yokogawa Electric Corporation | Paper machine controller for operating slices and method of controlling the same |
| WO1995006875A1 (en) * | 1993-09-03 | 1995-03-09 | Measurex Corporation | Automatic cross-directional control zone alignment for sheet-making systems |
| CN1190684A (en) * | 1996-08-30 | 1998-08-19 | 沃依特·苏尔泽制纸机械公司 | Controller with multiple sensors |
| US6086237A (en) * | 1995-12-13 | 2000-07-11 | Measurex Devron Inc. | Automated identification of web shrinkage and alignment parameters in sheet making machinery using a modeled actuator response profile |
| US20040221978A1 (en) * | 2003-05-09 | 2004-11-11 | Tran Peter Quang | Method and apparatus for controlling cross-machine direction (CD) controller settings to improve CD control performance in a web making machine |
| US20070039705A1 (en) * | 2005-08-22 | 2007-02-22 | Honeywell Asca Inc. | Reverse bump test for closed-loop identification of CD controller alignment |
| CA2410859C (en) * | 2000-06-13 | 2010-01-26 | Abb Inc. | Controlling cross machine profile in sheet making |
| US20110290438A1 (en) * | 2010-05-31 | 2011-12-01 | Honeywell Asca Inc. | Closed-loop Monitoring and Identification of CD Alignment for Papermaking Processes |
Family Cites Families (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5351184A (en) | 1993-01-26 | 1994-09-27 | Honeywell Inc. | Method of multivariable predictive control utilizing range control |
| US5539634A (en) | 1993-09-03 | 1996-07-23 | Measurex Corporation | Sheetmaking system identification using synthetic measurement produced from redundant noisy measurements |
| US5574638A (en) | 1995-04-03 | 1996-11-12 | Lu; Zhuxin J. | Method of optimal scaling of variables in a multivariable predictive controller utilizing range control |
| US5572420A (en) | 1995-04-03 | 1996-11-05 | Honeywell Inc. | Method of optimal controller design for multivariable predictive control utilizing range control |
| US5561599A (en) | 1995-06-14 | 1996-10-01 | Honeywell Inc. | Method of incorporating independent feedforward control in a multivariable predictive controller |
| US5892679A (en) | 1996-09-13 | 1999-04-06 | Honeywell-Measurex Corporation | Method and system for controlling a multiple input/output process with minimum latency using a pseudo inverse constant |
| US5853543A (en) | 1997-01-27 | 1998-12-29 | Honeywell-Measurex Corporation | Method for monitoring and controlling water content in paper stock in a paper making machine |
| US6092003A (en) | 1998-01-26 | 2000-07-18 | Honeywell-Measurex Corporation | Paper stock shear and formation control |
| US6080278A (en) | 1998-01-27 | 2000-06-27 | Honeywell-Measurex Corporation | Fast CD and MD control in a sheetmaking machine |
| US6149770A (en) | 1998-04-14 | 2000-11-21 | Honeywell-Measurex Corporation | Underwire water weight turbulence sensor |
| US6168687B1 (en) | 1998-04-24 | 2001-01-02 | Honeywell-Measurex Corporation | System and method for sheet measurement and control in papermaking machine |
| US6233495B1 (en) | 1998-06-12 | 2001-05-15 | Abb Automation, Inc. | Methods for modeling two-dimensional responses of cross-machine direction actuators in sheet-forming processes |
| FI107964B (en) | 1999-12-31 | 2001-10-31 | Metso Paper Automation Oy | Method and apparatus for identifying paper machine actuator alignment |
| US6650947B2 (en) | 2001-03-23 | 2003-11-18 | Metso Automation Oy | Multi-variable control loop assessment |
| US6805899B2 (en) | 2002-01-30 | 2004-10-19 | Honeywell International Inc. | Multi-measurement/sensor coating consolidation detection method and system |
| US6807510B1 (en) | 2003-05-05 | 2004-10-19 | Honeywell Acsa Inc. | Model predictive controller for coordinated cross direction and machine direction control |
| US7454253B2 (en) | 2006-03-30 | 2008-11-18 | Honeywell Asca Inc. | Fast performance prediction of multivariable model predictive controller for paper machine cross-directional processes |
| US7689296B2 (en) | 2006-04-28 | 2010-03-30 | Honeywell Asca Inc. | Apparatus and method for controlling a paper machine or other machine using measurement predictions based on asynchronous sensor information |
| US7577483B2 (en) | 2006-05-25 | 2009-08-18 | Honeywell Asca Inc. | Automatic tuning method for multivariable model predictive controllers |
-
2012
- 2012-03-28 US US13/433,101 patent/US9511969B2/en active Active
-
2013
- 2013-03-13 EP EP13768923.8A patent/EP2831341A4/en not_active Withdrawn
- 2013-03-13 WO PCT/CA2013/000241 patent/WO2013142960A1/en active Application Filing
- 2013-03-13 CA CA2867212A patent/CA2867212C/en active Active
- 2013-03-13 CN CN201380028046.8A patent/CN104350202A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4093988A (en) * | 1976-11-08 | 1978-06-06 | General Electric Company | High speed frequency response measurement |
| US5170357A (en) * | 1989-10-31 | 1992-12-08 | Yokogawa Electric Corporation | Paper machine controller for operating slices and method of controlling the same |
| WO1995006875A1 (en) * | 1993-09-03 | 1995-03-09 | Measurex Corporation | Automatic cross-directional control zone alignment for sheet-making systems |
| US6086237A (en) * | 1995-12-13 | 2000-07-11 | Measurex Devron Inc. | Automated identification of web shrinkage and alignment parameters in sheet making machinery using a modeled actuator response profile |
| CN1190684A (en) * | 1996-08-30 | 1998-08-19 | 沃依特·苏尔泽制纸机械公司 | Controller with multiple sensors |
| CA2410859C (en) * | 2000-06-13 | 2010-01-26 | Abb Inc. | Controlling cross machine profile in sheet making |
| US20040221978A1 (en) * | 2003-05-09 | 2004-11-11 | Tran Peter Quang | Method and apparatus for controlling cross-machine direction (CD) controller settings to improve CD control performance in a web making machine |
| US20070039705A1 (en) * | 2005-08-22 | 2007-02-22 | Honeywell Asca Inc. | Reverse bump test for closed-loop identification of CD controller alignment |
| US20110290438A1 (en) * | 2010-05-31 | 2011-12-01 | Honeywell Asca Inc. | Closed-loop Monitoring and Identification of CD Alignment for Papermaking Processes |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109208373A (en) * | 2017-07-06 | 2019-01-15 | 霍尼韦尔国际公司 | For enhancing the defect analysis of continuous coiled material sheet material, classification and the reparation of equipment efficiency and output |
| CN110227294A (en) * | 2019-06-17 | 2019-09-13 | 中国工程物理研究院激光聚变研究中心 | Polishing fluid circulated filter system |
| CN110227294B (en) * | 2019-06-17 | 2024-04-19 | 中国工程物理研究院激光聚变研究中心 | Polishing solution circulating and filtering system |
Also Published As
| Publication number | Publication date |
|---|---|
| CA2867212A1 (en) | 2013-10-03 |
| US9511969B2 (en) | 2016-12-06 |
| CA2867212C (en) | 2020-03-24 |
| US20130262019A1 (en) | 2013-10-03 |
| WO2013142960A1 (en) | 2013-10-03 |
| EP2831341A4 (en) | 2017-10-11 |
| EP2831341A1 (en) | 2015-02-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104350202A (en) | Closed-loop alignment identification with adaptive probing signal design technique for web manufacturing processing systems | |
| EP2576899B1 (en) | Closed-loop monitoring and identification of cd alignment for papermaking processes | |
| DaCosta et al. | Adaptive operator selection with dynamic multi-armed bandits | |
| DE102017104438B4 (en) | Machine learning device, laser device and machine learning method | |
| CN109195769A (en) | A kind of method of monitor production process, a kind of method of indirect derivation system relationship, a kind of method adjusting quality, a kind of method starting production process, a kind of method and a kind of equipment for manufacturing extruded product for manufacturing extruded product | |
| CN102325941B (en) | Method of creating generalized response model for actuator zones operable to control sheet properties and system thereof | |
| CN101490631A (en) | Fast performance prediction of multivariable model predictive controller for paper machine cross-directional processes | |
| WO1991014183A1 (en) | Actuation cell response and mapping determinations for web forming machines | |
| DE10212909A1 (en) | Multivariable control evaluation method e.g. for paper manufacture, involves using orthogonal disturbance method to locate orthogonal latent variables, which correlate with residual control circuit model | |
| CN106460727A (en) | Method and device for monitoring a parameter of a rocket engine | |
| EP3457226B1 (en) | Application of model predictive control (mpc)-based forced ramping of process input variables and process output reference trajectory design over a prediction horizon for mpc-based paper machine grade change control | |
| CN102325940B (en) | Method and apparatus for creating a comprehensive response model for a sheet forming machine | |
| EP3757795A1 (en) | Method and device for optimal distribution of test cases to different test platforms | |
| Zeydan | Modelling the woven fabric strength using artificial neural network and Taguchi methodologies | |
| US6584366B1 (en) | Method and arrangement for neural modeling of a paper winding device | |
| WO2021028111A1 (en) | Method and apparatus for optimizing an inductive heat-treatment process with the aid of a bayesian optimization process | |
| US20180087220A1 (en) | Method of Designing Model Predictive Control for Cross Directional Flat Sheet Manufacturing Processes to Guarantee Temporal Robust Stability and Performance | |
| CN107059180A (en) | Adjustment work platform and the yarn clearer on yarn manufacture weaving loom（Yarn qualities sensor）Method | |
| Wang et al. | Lyapunov-based controller design for bounded dynamic stochastic distribution control | |
| US10174456B2 (en) | Technique to improve paper machine cross-directional model predictive control performance by creating a measurement profile reference trajectory | |
| JP2020021301A (en) | Training data evaluation device, training data evaluation method, and program | |
| US6768972B1 (en) | Method and device for reducing a number of measured values of a technical system | |
| Bortolin | On modelling and estimation of curl and twist in multi-ply paperboard | |
| US20170045419A1 (en) | System and method for step test-free machine modeling using statistical information about multiple web manufacturing or processing systems | |
| Wang | Detect unexpected changes of particle size distribution in paper-making white water systems |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150211 |